The Persistent Power of Behavioral Change: Long-Run Impacts of Temporary Savings Subsidies for the Poor”, American Economic Journal: Applied Economics, 2018, 10(3), pp. 67-100.

Male Social Status and Women’s Work” (with Arielle Bernhardt, Erica Field, Rohini Pande, Natalia Rigol, and Charity Troyer-Moore). AEA Papers and Proceedings, 2018, 108, pp. 363-367.

Salary Delays and Overdrafts in Rural Ghana” (with Niklas Buehren, Virginia Ceretti, Ervin Dervisevic, Markus Goldstein, Leora Klapper, and Tricia Koroknay-Palicz). AEA Papers and Proceedings, 2018, 108, pp. 449-452.

The Cost of Convenience? Transaction Costs, Bargaining Power, and Savings Account Use in Kenya”. Journal of Human Resources, 2017, 52(4), pp. 919-945.

Do Opposites Detract? Intrahousehold Preference Heterogeneity and Inefficient Strategic Savings”. American Economic Journal: Applied Economics, 2015, 7(2), pp. 135-174.

Price Subsidies, Diagnostic Tests, and Targeting of Malaria Treatment: Evidence From a Randomized Controlled Trial” (with Pascaline Dupas and Jessica Cohen). American Economic Review, 2015, 105(2), pp. 609-645.

Working Papers

"The Contribution of Patients and Providers to the Overuse of Prescription Drugs" (with Carolina Lopez and Anja Sautmann). November 2018. NBER Working Paper No. 25284. Under Review.

"The Limits of Commitment: Who Benefits from Illiquid Savings Products?" (with Niklas Buehren, Markus Goldstein, Leora Klapper, and Tricia Koroknay-Palicz). October 2018. World Bank Policy Research Working Paper No. 8633. Under Review.

"On Her Account: Can Strengthening Women's Financial Control Boost Female Labor Supply?" (with Erica Field, Rohini Pande, Natalia Rigol, and Charity Troyer Moore). November 2016. New version coming soon!

Female Labor Force Participation in Asia: Indonesia Country Study” (with Smita Das), February 2016. ADB Economics Working Paper No. 474.

Selected Work in Progress

“Empowering Indonesian Migrant Workers to Access Quality Overseas Placement Services” (with Samuel Bazzi, Lisa Cameron, and Firman Witoelar).

Status: Baseline survey and intervention completed. Waves 1 and 2 follow-up data collection completed.

Abstract: Roughly 500,000 Indonesian workers migrate abroad on temporary work contracts each year. A critical determinant of a worker’s migration experience is his or her choice of placement agency, which is responsible for pre-departure training and facilitating a worker-employee match. In theory, migrants have a great deal of choice when selecting an agency: roughly 500 agencies are legally registered with the Indonesian government and the median village in our study had 19 agencies place female workers in 2015. At the same time, there is a great deal of heterogeneity in agency quality, even conditional on observed demographic characteristics of workers. One explanation for this apparent heterogeneity in the face of competition is that migrants lack the information needed to select a high-quality agency. Alternatively, even if migrants have access to information, they may not see the value in aggregating and attending to it.

We conducted a field experiment in 400 Indonesian villages to assess how (lack of) information and migrant beliefs interact to influence migration decisions. The experiment included one arm where potential migrants were given agency “report cards”, which aggregated novel information about agency quality that we collected through an extensive baseline survey. Another arm provided no quality information, but instead educated potential migrants about the “theory of change” for why choosing a high-quality agency is important. The final treatment arm combined the theory-of-change education with agency report cards. We link our experimental villages to detailed administrative placement data, which will allow us to assess how the interventions changed the structure of the village-level migration market, both in terms of number of workers placed and the identity of the agencies. These data are complemented by endline surveys of potential and actual migrants, which will allow us to assess the impact of the interventions on individual beliefs and welfare.

AEA RCT Registry: click here

“Following Up for Better Health: Improving Non-Communicable Disease Compliance in Urban India” (with Daniel Bennett).

Status: Wave 1 follow-up data collection completed.

Abstract: The majority of individuals with non-communicable diseases (NCDs) like diabetes live in the developing world, where prevalence rates are growing rapidly as nations become richer and more urban. In cities, where most individuals have access to a variety of medical providers, improving adherence to treatment is key to reducing morbidity from NCDs. This project will evaluate the role that different behavioral factors (particularly psychic costs of seeking treatment, limited attention, and loss aversion) play in limiting diabetes adherence. The research design consists of a randomized controlled trial with three sets of cross-cutting interventions. The first set of interventions will provide patients with information about diabetes and the importance of adherence to treatment. Here, patients will be randomly selected to receive either negative/loss-framed information or positive/gain-framed information. The second intervention is a voice-based mobile phone reminder service. We will deploy reminders via a novel two-stage design– a selective trial in the parlance of Chassang et al. (2012) – which will allow us to estimate both willingness to pay for reminders and treatment effects conditional on willingness to pay.

AEA RCT Registry: click here

“Building Transparency and Trust in the Financial System with Voice Notifications” (with Erica Field, Rohini Pande, Natalia Rigol, and Charity Troyer Moore).

Status: In the field.

Abstract: The Government of India’s ambitious Pradhan Mantri Jan Dhan Yojana (PMJDY) scheme has opened over 226 million new bank accounts since its inception in August 2014, with the aim of drawing India’s unbanked rural poor into the formal financial system. Although this is an impressive achievement, many of these accounts are dormant, or are only used to receive government benefits. Thus, significant barriers to financial inclusion remain: Low financial literacy rates, especially among women, result in discomfort with new technologies used by last-mile banking kiosks, such as fingerprint readers. Moreover, banking kiosk operators often hold an informational monopoly on account details, as illiterate individuals cannot read transaction receipts. Hence, the rural poor still lack a transparent, verifiable, easily understood way to access information about account balances and the timing of direct deposit transfers. Given the deep mobile phone penetration in rural India and the fact that even low-literacy individuals regularly use the basic calling functions of mobile phones, can mobile technology facilitate access to and improve the quality of financial services for the rural poor? In this project, we will partner with a large public Indian bank to pilot an interactive voice response (IVR) system to give low-income, low-financial-literacy individuals information on account transactions and government benefits receipt. If the pilot is successful, we plan to scale the project to assess the impact of IVR notifications on bank account use, financial inclusion, and access to social protection benefits delivered via bank accounts.

AEA RCT Registry: click here

“Unpacking the Demand for Commitment” (with Markus Goldstein, Leora Klapper, and Tricia Koroknay-Palicz).

Status: In the field.


This study is designed to increase understanding of how intra-household dynamics and gender impact savings, and particular the demand for illiquid savings accounts. As part of this study, we offer mobile-phone based bank savings accounts that are linked to mobile money to married men and women in Ghana’s Volta region. The design randomly vary three elements: (i) whether a liquidity restriction on the account is binding; (ii) spousal information regarding which liquidity restriction is in place; and (iii) the value of incentive payments for choosing a binding versus non-binding liquidity restrictions. This design allows us to (a) map the demand curve for hard vs soft commitment; (b) assess the extent to which spousal information impacts demand for hard vs soft commitment; and (c) assess how demand varies by gender.

AEA RCT Registry: click here